Bitcoin Forum

I'm working on a distributed NP-Complete problem solver, which I will post about soon.

What do you think would happen to Bitcoin if people knew there was a Bitcoin address hacker out there?  All it takes is 100,000 people working together...

I think someone on Reddit did the maths, and it is practically impossible to crack someone's keys with current hardware (unless you maybe have access to one of the best computers in the world). I don't even think a large bot net would be enough.

Not a bot net.  An altcoin.

But more importantly, what do you think would happen to Bitcoin if there was a network out there breaking into 1 address/week or so?

Probably people would split up their money over more addresses, I guess.

*Hint*
Major pools can combines their computing power to brute force private keys...lalala

Sure.  Google could probably do it, along with the US government. ;D

I do not know the maths off of the top of my head, but there is plenty of it out there if you care to look, but it basically comes down to the chances, even with the number of people mining an "alt coin" of finding an address that has been used would take millions if not billions of years to brute force, so, fortunately for the bitcoin network, it is not possible to find one every week.

No, 100,000 people is not enought, in fact nothing is enought, so strong the private keys are to bruteforce

Unless you have wallet.dat+keylogger active on fellow bitcoiner computer

I would be using an algorithmic method of cracking a private key, not brute forcing it, if that matters.

Anyway, I really just want to make an alt-coin for my own purposes.  I'm just worried that my alt-coin will break Bitcoin.

Nothing would happen.  You could have a trillion people/computers working together and it wouldn't make any significant difference.

Well if everyone's so certain that an absolutely huge network of computers all working together still can't break into a bitcoin address, then I'll go ahead and try to make my altcoin.

Let's do some math. The keyspace you have to search to loot a bitcoin address is 160 bits. Let's see, how much time you have to spend to have even a one-in-a-thousand chance of being successful.

Total keyspace:
160 bits.

Suppose you're lucky. You find your key while searching 1/1024 (2^10) of the keyspace.
150 bits left.

You have a device that can guess at a rate of about 1 billion (2^30) per second.
120 bits left.

You have one million (2^20) of these devices.
100 bits left.

So, you'll need 2^100 seconds to guess the key.

That's about 3 trillion times the age of the universe.

I'm not brute forcing the key, I'm algorithmically solving for the key.  But I don't think that changes the math at all.

OK, even if I got all the computers in the world to work together and not duplicate work, it would still probably be impossible to break into a single address.  That's kind of comforting, but now I have nothing cool to do with my altcoin.

Those words in that context are complete nonsense.

There is no known method of deriving the digest from a hash or the private key from the pubkey other than brute force (unless the algorithm is cryptographically broken and RIPEMD-160, SHA-256, and ECDSA are not).  That is the entire point of hashing algorithms and public key cryptography.   If there was a faster method than brute force the algorithms would be broken.

Not sure what you think brute force means but baring a cryptographic break or quantum computer capable of Shor's algorithm (limited to attacks on known PubKeys only) your options are:
a) brute force
b) see option a.

Get ready to become very famous!

Brute forcing a private key means to check every possible private key to see if it's the correct one.  That is not what my network would do, but what my network would do is really no faster.

But I've realized an entirely separate problem with my approach: It lets me solve hard problems, but only about 1million times faster than my current computer. Which still is not fast enough to solve NP-complete problems quickly.